Skip to content

GPU ROADMAP 2026–2028

Release timeline for next-generation data center GPUs from NVIDIA, AMD, Google, and Intel. Updated as new information becomes available.

AvailableSamplingAnnouncedRumored

2026

8 GPUs

NVIDIA B300 Ultra

NVIDIAAvailable
Q1 2026

Process

TSMC 4NP

Memory

288GB HBM3e

FP8 TFLOPS

7,000

TDP

1,400W

Key Feature

2× H100 per GPU, 10 TB/s memory bandwidth

Blackwell Ultra generation. Shipping in limited quantity to hyperscalers and cloud providers.

AMD Instinct MI350X

AMDAvailable
Q1 2026

Process

TSMC 3nm

Memory

288GB HBM3e

FP8 TFLOPS

4,600

TDP

1,000W

Key Feature

CDNA4 architecture, 8 TB/s memory bandwidth

AMD's CDNA4 generation. Competitive with B200 on VRAM; still ramps to availability.

NVIDIA B300A

NVIDIAAnnounced
Q2 2026

Process

TSMC 4NP

Memory

288GB HBM3e (unified CPU+GPU)

FP8 TFLOPS

TBD

TDP

~1,200W

Key Feature

ARM Neoverse CPU + Blackwell GPU die — full unified memory

Grace Blackwell Ultra variant. Targeting workloads that benefit from unified CPU-GPU memory addressing.

Google TPU v7

GOOGLEAvailable
Q1 2026

Process

Custom ASIC

Memory

TBD

FP8 TFLOPS

~4,000 (est.)

TDP

~600W

Key Feature

2× TPU v5p performance, JAX-native, pods to 8,192 chips

Available on Google Cloud. Best for JAX-native workloads and Google-style distributed training.

AMD Instinct MI355X

AMDAvailable
Q2 2026

Process

TSMC 3nm

Memory

288GB HBM3e

FP8 TFLOPS

4,610

TDP

1,400W

Key Feature

Highest VRAM of any commercially available GPU

Premium CDNA4 variant. Maximum VRAM configuration for very large model training.

AMD Instinct MI400X

AMDAnnounced
Q3 2026

Process

TSMC 3nm+

Memory

384GB+ HBM4 (est.)

FP8 TFLOPS

~8,000 (est.)

TDP

~1,200W (est.)

Key Feature

CDNA5 architecture, HBM4 memory, Infinity Fabric 4.0

Next-generation CDNA5 architecture. AMD has confirmed MI400 series for late 2026. HBM4 expected.

NVIDIA Rubin (R100)

NVIDIAAnnounced
Q4 2026

Process

TSMC 3nm (est.)

Memory

TBD (HBM4e est.)

FP8 TFLOPS

~10,000+ (est.)

TDP

TBD

Key Feature

Post-Blackwell architecture, NVLink 6.0

NVIDIA confirmed Rubin at GTC 2025. Expected to ship to select partners in late 2026, broad availability 2027.

Intel Falcon Shores

INTELAnnounced
Q4 2026

Process

Intel 18A

Memory

TBD (HBM3e est.)

FP8 TFLOPS

TBD

TDP

TBD

Key Feature

Integrated Xe GPU + compute tiles, Gaudi successor

Intel's next-gen AI accelerator replacing Gaudi 3. Targeting hyperscale and enterprise AI deployments.

2027

3 GPUs

Google TPU v8

GOOGLERumored
Q2 2027

Process

Custom ASIC

Memory

TBD

FP8 TFLOPS

TBD

TDP

TBD

Key Feature

Liquid cooling, 2× TPU v7 performance target

Expected to maintain Google's ~2-year generation cadence. Likely announced at Google Next 2027.

NVIDIA Rubin Ultra (RU100)

NVIDIARumored
Q1 2027

Process

TSMC 3nm

Memory

TBD (HBM4e)

FP8 TFLOPS

~14,000 (est.)

TDP

TBD

Key Feature

Rubin Ultra refresh — follows Blackwell → Blackwell Ultra pattern

Based on NVIDIA's Ultra refresh cadence (H100→H200, B200→B300). Expected 2–3 quarters after base Rubin.

AMD Instinct MI450X

AMDRumored
Q2 2027

Process

TSMC 2nm (est.)

Memory

TBD (HBM4e est.)

FP8 TFLOPS

TBD

TDP

TBD

Key Feature

CDNA6 architecture target

AMD's roadmap targets annual GPU architecture updates. MI450X expected to compete with Rubin Ultra.

Disclaimer

Specs for "Announced" and "Rumored" GPUs are estimates based on official announcements, roadmap slides, and reputable industry sources. Actual specs, pricing, and availability dates may differ significantly. GPUAdvisor updates this page as new information becomes available.

Roadmap Alerts

Get notified when new GPU specs are announced or release dates are confirmed. No spam.